Critical Reviews in Analytical Chemistry, 36:3–11, 2006Copyright c© Taylor and Francis Group, LLCISSN: 1040-8347 print / 1547-6510 onlineDOI: 10.1080/10408340500451957

REVIEW

Analytical Methods for the Determinationof Organic Acids in Honey

Ines Mato, Jose F. Huidobro, and Jesus Simal-LozanoFacultad de Farmacia, Departamento de Quımica Analıtica, Nutricion y Bromatologıa, Area deNutricion y Bromatologıa, Universidad de Santiago, Galicia, Spain

M. Teresa SanchoFacultad de Ciencias, Departamento de Biotecnologıa y Ciencia de los Alimentos, Area de Nutricion yBromatologıa, Universidad de Burgos, Burgos, Castilla y Leon, Spain

Although organic acids represent less than 0.5% of honey’s constituents, they make importantcontributions to organoleptic, physical, and chemical properties of honey. They could be usedas fermentation indicators, for the treatment of Varroa infestation, and to discriminate amonghoneys according to their botanical and/or geographical origins. This article reviews the currentliterature related to the analytical methods (enzymatic, chromatographic and electrophoretic)that have been applied recently to the determination of honey’s organic acids. The advantagesand disadvantages of all the procedures described are also discussed. This review has beenwritten to make the study of these interesting honey component easier.

Keywords organic acids, honey, analytical method review

Although honey is comprised of organic acids in a smallproportion (they only amount to less than 0.5%), the study ofthese components is interesting because they make importantcontributions to both its organoleptic properties, such as colorand flavor, and to its physical and chemical properties, suchas pH, acidity and electrical conductivity (1, 2). They also haveantibacterial and anti-oxidant activities; together with other sub-stances (3–5), they could be used as fermentation indicators, orfor the treatment of Varroa infestation (6–8). They could alsobe deployed to discriminate between honeys according to theirbotanical and/or geographical origins (9–12).

Despite their increasing importance in honey, there is rela-tively little information available about these organic acids andtheir analysis. In a previous article, the current literature relatedto the significance of non-aromatic organic acids in honey wasreviewed (13). Therefore, we have also decided to commentthe current analytical methods that allow the determination ofhoney organic acids. In this review, enzymatic, chromatographic

Address correspondence to Jose F. Huidobro, Facultad de Farma-cia, Departamento de Quımica Analıtica, Nutricion y Bromatologıa,Area de Nutricion y Bromatologıa, Universidad de Santiago, 15782Santiago de Compostela, Galicia, Spain. Fax: +34 81 594612. E-mail:qnhuidob@usc.es

and electrophoretic methods have been presented to identify andquantify the most important organic acids in honey, individuallyor as a group simultaneously. The advantages and disadvantagesof all the procedures described are also discussed.

ENZYMATIC METHODSEnzymatic methods have been used, mainly, for the quantifi-

cation of citric, malic, formic, D- and L-lactic, oxalic and totalD-gluconic acids in honeys. However, it would be possible forthe determination of other acids such as acetic, L-ascorbic andsuccinic acids (14).

All enzymatic methods are based on the measurement of theincrease or decrease in absorbance of the coenzymes NADH(nicotinamide-adenine dinucleotide, reduced form) or NADPH(nicotinamide-adenine dinucleotide phosphate, reduced form),which absorb light in the long wavelength region, generally at340 nm.

Tourn et al. (15) applied enzymatic tests, without previousclarification, to quantify citric and malic acids in 5 samples ofhoney (3 of honeydew, and 2 of multifloral honeys) and 1 sam-ple of royal jelly. According to these authors, the enzymaticprocedure showed noticeable advantages with respect to othermethods, because of its specificity, sensitivity, rapidity and sim-plicity. These authors provided neither data of precision nor ofrecovery of their proposed enzymatic procedure.

3

4 I. MATO ET AL.

Stoya et al. (16) determined formic acid content by applyingan enzymatic analysis to honeys coming from hives untreatedagainst varroase (natural content of formic acid in honey) andto honeys coming from hives that had been treated with formicacid against Varroa infestation (residual formic acid, after treat-ment). In this paper, only data of recovery are shown, with anaverage recovery percentage of 96.0 ± 6.5%. These researchersconcluded that honeys from treated beehives contained from 54to 82 times more formic acid than honeys from untreated bee-hives. Those levels of formic acid decreased asymptotically withtime, so that between 5 to 8 months after the treatment, the levelsof formic acid in honeys from treated beehives were more or lesssimilar to those found in honeys from untreated beehives.

Stoya et al. (17) determined D-(−) and L-(+)-lactic acid con-tents by using an enzymatic method both in honeys untreatedagainst varroase (natural content of these acids) and in hon-eys coming from hives that had been treated with D-(−) andL-(+)-lactic acids against Varroa infestation (residual acids, af-ter treatment). In this paper quotients between both isomers(D-(−)/L-(+)) are shown for each sample. In the vast majorityof samples, the mentioned quotient was lower than 3, althoughthere were honeys with a quotient higher than 20. Average recov-ery percentage for D-(−)-lactic acid was of 83 ± 13%, and forL-(+)-lactic acid was of 96 ± 6%. These researchers found thatin honeys coming from treated hives, only L-(+)-lactic acid con-tents were increased 10 times in comparison with normal levels.In contrast, D-(−)-lactic acid levels in honeys from treated anduntreated beehives were similar. About 7 or 8 weeks after treat-ment, L-(+)-lactic acid levels decreased asymptotically to usuallevels.

Hansen and Guldborg (18) analyzed formic acid contents ofDanish honeys. The detection limit was 5–10 mg formic acid/kghoney. No additional validation data was shown. Some sam-ples came from hives untreated against varroase, and other sam-ples came from hives that did not suffer from varroase but weretreated with formic acid, and harvested 8–9 months after treat-ment. These researchers concluded that after this time, formicacid levels in honeys from treated beehives were similar to thenatural levels in honeys from untreated beehives.

Talpay (19) quantified citric acid contents in 153 floral hon-eys, 97 honeydew honeys, and 18 samples coming from bees thathad been artificially fed with sugar. The enzymatic test was ap-plied directly, obtaining a precision with coefficients of variation(RSD %) between 1.61% and 1.72% and a recovery percentageof 92.0 ± 6.8%. In this work, the author suggested that maybecitric acid content might be useful to distinguish between floral,honeydew honeys, and sugar-fed honeys. Citric acid levels weresignificantly lower in floral honeys than in honeydew honeys.However, no differences were found between honeys from floralorigin and from artificial sugar-fed origin.

Talpay (20) determined formic acid content of 306 honeysfrom different botanic and geographic origins. The enzymatictest that was used showed a precision with a coefficient of varia-tion (RSD %) of 4.09%, and a percentage of recovery of 97.6 ±

1.7%. This author concluded that chestnut (Castanea sativa),eucalyptus (Eucalyptus sp.), ling heather (Calluna sp.), andbell heather (Erica sp.) honeys, in this decreasing order, weremarkedly different from honeydew and other floral honeys inrelation to the formic acid content. No statistical analysis wasperformed.

Sabatini et al. (21) employed enzymatic tests to determineformic, D-lactic and L-lactic acids contents in 42 unifloral honeysfrom 7 different botanical sources. These authors highlightedthat the honey solution was previously clarified with activatedcharcoal to minimize the influence of honey color and to removecertain instability of the spectrophotometric measurement. Noresult of both precision and recovery was given. The conclusionof this work was that, with regard to formic acid contents, therewere important differences between light and dark honeys.

Mato et al. (22) set up an enzymatic micro-method to de-termine total D-gluconic acid in honey, which includes freeD-gluconic acid and D-glucono-δ-lactone, because the latter hasnot been able to be quantified separately. To be sure of the com-plete decomposition of D-glucono-δ-lactone, these authors tookhoney solution to a pH of 10.5 for 10 minutes at room tem-perature. Afterwards, the pH was dropped to 7.8, which is theoptimum pH for enzymatic activity. This method combines pre-cision (RSD% ≤0.30%), good recovery (99.8 ± 0.40%), speci-ficity, simplicity and low cost. Gheldof et al. (5) have appliedthis method for studying the relationship between anti-oxidantactivity and gluconic acid (as an indicator of organic acid oforganic acid concentration). In this paper, no significant corre-lation between gluconic acid level and anti-oxidant activity wasdetected. This method was also used by Pulcini et al. (23), whoanalysed total D-gluconic acid of 112 Italian unifloral honeys.In these honeys, the precision of results (RSD%) were between0.59% and 3.93% and mean recovery was 97.9 ± 1.12%).

Mutinelli et al. (7) determined honey’s oxalic acid contentby using an enzymatic method. Detection limit was 3 mg oxalicacid/kg honey. No additional validation data were shown. In thiswork, the researchers analysed the levels of oxalic acid in honeysfrom untreated beehives and from treated beehives against Var-roa infestation with oxalic acid. No statistical differences werefound between pre- and post-treatment content of oxalic acid.

Mato et al. (24) applied an enzymatic micro-method for thedirect determination of L-malic acid of honey, studying bothprecision and accuracy. RSD% was ≤3.5% and the average re-covery was 100 ± 3.5%. Furthermore, these authors carriedout several assays for the clarification of the samples by usingdifferent clarifying agents: polyvinylpolypyrrolidone [PVPP],Carrez, Carrez with NaOH, Carrez with KOH, Carrez withpolyvinylpolypyrrolidone [PVPP], and activated charcoal, butno improvement of precision and recovery was observed.

Mato et al. (25) developed an enzymatic specific micro-method for honey’s citric acid, by using an UV-test with pre-vious clarification of the sample with polyvinylpolypyrrolidone(PVPP). In this work, authors compared precision and accuracyof the proposed procedure (with clarification) in relation to those

METHODS OF ORGANIC ACIDS IN HONEY 5

of the direct analysis, that is to say, without any clarification. Au-thors concluded that precision and accuracy of the method withPVPP clarification (precision between 0.26 and 1.60%, and re-covery between 98.0 and 100.9%) were better than the resultsof the direct method (precision between 1.02 and 2.66%, andrecovery between 84.0 and 115.6%).

Cossu and Alamanni (26) adapted to honey an enzymaticmethod widely used for the analysis of oxalate in urine, by de-termining oxalic acid in 40 Sardinian (Italy) honeys. Sampleswere previously clarified with activated charcoal. The detectionlimit for oxalic acid was 0.40 mg/L and the quantification limitwas 1.33 mg/L. Average recovery percentage was 90.8 ± 1.0%.Alamanni et al. (27) analyzed oxalic acid in 49 floral honeysby applying the enzymatic method set up by Cossu and Ala-manni (26). Results were compared with those obtained by us-ing a HPLC method set up by these authors for the simultaneousdetermination of oxalic, formic and lactic acids with previoussolid-phase extraction (SPE) of the samples’ acids. Alamanniet al. (27) claimed that there was a good correlation betweenthe results obtained by both methods, enzymatic and chromato-graphic. The authors specified that enzymatic quantification ofoxalic acid in tree strawberry (Arbutus unedo) honeys was notpossible due to the fact that there was a parallel uncontrollablereaction, which prevented absorbance stabilization.

Bogdanov et al. (28) determined formic and oxalic acid con-tents in honey by applying the enzymatic test used for oxalicacid analysis, which allows quantifying both acids in the samesample. Coefficients of variation (RSD %) ranged between 2.9and 4.4% for formic acid, and between 13.4 and 18.4% for ox-alic acid. Recovery for formic acid was of 93 ± 4%, whereas foroxalic acid was of 84 ± 24%. Specific detection limits were notshown for both determinations, because authors had pointed outthat these limits ranged between 3 and 5 mg/kg. In this work,Bogdanov et al. (28) found a small, but significant, increase offormic acid in honey after a 7-day treatment of hives with formicacid. Increases were similar throughout the different years ofanalysis, so that the authors thought that formic acid remains inthe sugar feed in the brood combs. According to these authors,these residues were not problematic, because they had no influ-ence on the quality of honey. On the contrary, they found thatoxalic acid content remained unchanged, even after two succes-sive treatments during the same autumn. No rise of free aciditywas found after combining a treatment with formic and oxalicacids throughout the 3 years of assays.

The main advantages of enzymatic methods include highspecificity that even allows the determination of D/L isomersof some organic acids. In addition, the equipment and materialsnecessary for enzymatic analysis are very simple and availablein the vast majority of honey quality control laboratories. Enzy-matic procedures could also be considered excellent as referencemethods, but in this case, more precision and accuracy studieswould be necessary, because several researchers point out noth-ing about these studies. And, the precision and recovery resultsof other authors are not good enough as they would be expected

with enzymatic methods. The main disadvantage of enzymatictests is that they allow the determination of only one acid, sothat if an organic acid profile of a given honey is necessary tocharacterize the sample botanically or geographically, the enzy-matic analysis would be tedious and time consuming. Anotherdisadvantage is the stability of these kits, which is not very long,plus the assays must be carried out in a few days.

CHROMATOGRAPHIC METHODSThe chromatographic methods developed for the determina-

tion of organic acids in honey samples have been divided intothree groups: paper and on-column ion exchange chromatogra-phy, gas chromatography and liquid chromatography.

Paper and On-Column Ion Exchange ChromatographyStinson et al. (29) first reported a separation and quantifica-

tion of several organic acids simultaneously in honey samples(butyric, acetic, formic, lactic, succinic, pyroglutamic, malic,citric and gluconic acids) by using chromatographic methods aspaper or on-column ion exchange chromatography.

Gas Chromatography (GC)Speer and Montag (30, 31) quantified phenylacetic and ben-

zoic acids in 32 honeys of different botanical origin by usingGC-FID. These acids were extracted from honey with ethyl ac-etate. The method recoveries were 85.9% for phenylacetic acidand 97% for benzoic acid. Heather honeys contained the higheramounts of both phenylacetic and benzoic acids.

Steeg and Montag (32–34) determined the content of 24aromatic organic acids as free acids in honeys from differ-ent floral sources. These acids as their trimethylsilyl deriva-tives were analyzed by a GC-MS method after extraction withethylacetate. Benzoic, β-phenylactic, 4-hydroxybenzoic and4-hydroxyphenylacetic acids were found in all samples. Theauthors found differences in the content of the aromatic or-ganic acids according to the floral source of honey. For example,heather honeys could be identified by the presence of a high con-centration of benzoic acid, phenylacetic acid, mandelic acid andβ-phenylactic acids. The differentiation of honeydew and flo-ral honeys were also possible because of the difference in theconcentration of protocatechuic acid.

Echigo and Takenaka (35) used a gas-chromatographic (GC)method for the determination of 10 honey organic acids: oxalic,malonic, succinic, fumaric, malic, α-ketoglutaric, tartaric, cis-aconitic, citric and gluconic acids in about 45 minutes. Becauseof the fact that these acids are hardly volatile, it was necessaryto obtain their trimethylsilyl derivatives.

Wilkins et al. (10) identified 32 honey dicarboxylic aliphaticacids as methyl esters, analyzing them by gas-chromatographyand mass spectrometry detection (GC-MS). Then, 18 acids werequantified, among them malic and succinic acids. Methodology,setting up extraction, derivatization and chromatographic anal-ysis had been previously described by Tan et al. (36) for the

6 I. MATO ET AL.

determination of different compounds, such as hydro-carbons (C21-C33), aromatic acids, diacids and linearchain fatty acids (C8-C28). Two aliphatic dicarboxylicacids (2-methoxybutanedioic and 4-hydroxy-3-methyl-trans-2-pentenedioic) were identified in Knightea excelsa honeys. Afteranalyzing more than 200 New Zealand honeys from differentbotanical origins, these two acids were only found in honeyswith an important contribution of pollen from Knightea excelsa,so that according to the researchers, 2-methoxybutanedioic and4-hydroxy-3-methyl-trans-2-pentenedioic acids could be con-sidered as markers of Knightea excelsa honeys.

Some methods, originally developed for the determina-tion of carbohydrates in honeys, can be used for the analy-sis of some acids. Horvath and Molnar-Perl (37) developeda GC-MS procedure for the simultaneous determination, astrimethylsilyl derivatives, of major honey constituents (sug-ars) and minority compounds (such as organic acids).Thismethod allowed quantifying 5 aliphatic carboxylic acids (malic,citric, isocitric, succinic and quinic acids) and 4 aromaticacids (benzoic, 4-hydroxibenzoic, 2,4-dihydroxibenzoic and2,6-dihydroxibenzoic acids). Sanz et al. (38) determined quinicor gluconic acids by using a GC-FID and GC-MS methods. Aderivatization process was also necessary for the detection ofthese acids which was analysed as trimethylsilyl derivatives.

Verzera et al. (39) set up a GC-MS method with previous solidphase microextraction (SPME) that allowed the identification of113 volatile compounds in 15 honeys from different botanicalsources coming from Sicily (Italy). Aliphatic organic acids, suchas acetic acid, and aromatic acids, such as benzoic acid, werefound among the volatile compounds. According to these re-searchers, their SPME-GC-MS method was simple, rapid, easilyreproduced, and provided useful information for the determina-tion of the floral source of honey. However, in this paper, noresult about the quantification of these compounds was given.

Pilz-Guther and Speer (40) developed a GC-FID to analyzedifferent organic acids like D,L-lactic, citric, succinic acid, andL-malic acids simultaneously in honey after clean-up by a SAX-cartridge. As occurs in others GC methods, it was necessary tocarry out a derivatization process. Citric and malic acid concen-trations obtained by this method are compared with the resultsobtained by the enzymatic method, and there were not statisti-cally significant differences between both groups.

Gas chromatography is greatly adequate to determine volatileorganic acids, such as aromatic acids, which are identified to-gether with other compounds of honey flavor. Nevertheless,when the analysis of aliphatic organic acids is necessary, gaschromatography does not seem to be the most suitable tech-nique, because the vast majority of these acids are not volatile,thus, making it necessary to derivatize them previously.

Liquid Chromatography (HPLC, IC)Perez-Cerrada et al. (41) developed a method to determine

inorganic anions such as chloride, phosphate and sulphate byionic chromatography (IC) with conductivity detection that also

allowed determining malic, tartaric and oxalic acids. This tech-nique was applied to food samples, such as cane sugar and di-etetic vegetable sugar, honey, lemon syrup, condensed milk andliquid caramel, among others. No matrix interferences were ob-served and the reproducibility, according to these authors, wasacceptable. Validation data for these organic acids determinationwere not given.

Jorg and Sontag (42) developed a HPLC method forqualitative and quantitative determination of phenolic acids(2-hydroxybenzoic, 3-hydroxybenzoic, 4-hydroxybenzoic,3,4-hydroxybenzoic, 4-hydroxycinnamic and 4-hydroxy-3-methoxycinnamic acid) in 11 honeys from different botanicalorigin. The phenolic acids were previously extracted fromhoney using ethyl acetate. The chromatographic separationwas performed in a reversed phase column using a mixture ofmethanol, glacial acetic acid and bidistilled water at pH of 3.0 asmobile phase. These acids were detected electrochemically us-ing a coulometric dual electrode detector, so they were oxidizedin the first cell and then reduced in the second one. Detectionlimits ranged between 0.3 to 1.5 ng for the oxidative modeand from 0.1 to 1.6 ng for the reductive detection mode. Therecoveries ranged from 87.8 and 97.8%, with a relative standarddeviation between 2.0 and 6.8%. No result about methodprecision was shown in this work. The authors concluded thatdistribution pattern of phenolic acids allows to differentiatebetween some honeys of different botanical origin, for example,3,4-hydroxybenzoic was found in large amount in honeydewhoneys. Therefore, further investigations would be necessary.

Cherchi et al. (9) identified several organic acids in honeyand quantified gluconic, pyruvic, malic, citric, succinic and fu-maric acids in 48 floral honeys by using a HPLC method withUV detection. Samples were previously purified by solid phaseextraction (SPE) with an ionic exchange cartridge. Chromato-graphic separation was carried out in 60 minutes by employ-ing two reversed-phase C18 columns connected in series, withsulphuric acid (pH = 2.45), at a flow rate of 0.7 mL/min asmobile phase. The detection limits ranged between 0.002 to3 ppm (w/w). Average recovery percentages ranged from 89and 104%, depending on the organic acid. In this paper, pre-cision data were not given. According to the authors, the de-scribed method would allow the simultaneous determination ofa number of organic acids, assisted by the fact that the prob-lem of overlapping peaks could be overcome by varying themobile phase pH. Polycarboxylic acids such as citric and fu-maric acids might be more sensitive to any change regardingthe mobile phase. Cherchi et al. (43) quantified gluconic, pyru-vic, malic, citric, succinic and fumaric acid contents of 37 uni-floral Sardinian (Italy) honeys, purifying the samples by solidphase extraction (SPE) and analyzing them by HPLC. They fol-lowed the method of Cherchi et al. (9). Furthermore, in order tocharacterize those Sardinian samples, these authors determinedother physical, chemical, organoleptic, microbiological and pa-lynological parameters. All these parameters were subdividedinto two groups: “Characterization parameters” and “Quality

METHODS OF ORGANIC ACIDS IN HONEY 7

Control parameters.” Organic acids and gluconic acid in partic-ular, were included in characterization parameters by Cherchiet al. (43). Serra-Bonhevı et al. (44) applied, with some modi-fications, the Cherchi et al. (9) method to determine 12 organicacids (as gluconic, oxalic, pyruvic, malic, citric, succinic, fu-maric, propionic, 2-oxopentanoic, glutaric, isobutyric, butyricacids) in broom honey (Spartocytisus supranubius L.) producedin The Canary Islands (Spain). In this work, the samples were de-termined directly without a clean-up procedure with cartridgesor other extraction procedure. Two columns connected in se-ries are used for performing the acid separation. Besides, dataof method validation (detection and quantification limits, re-coveries and precision) is shown. In the summary, the authorssuggested that malic acid can be used as a marker for the dif-ferentiation of broom honey, but any statistical treatment wasapply to confirm this idea.

Defilippi et al. (45) determined formic acid by ionic chro-matography (IC) with conductivity detector, using as mobilephase carbonate/bicarbonate buffer 0.25:2.75 mM (pH = 10.1),at a flow rate of 1 mL/min. Sample preparation was a simpledilution, because neither purification nor honey extraction werenecessary. Analysis time was shorter than 5 minutes. The detec-tion limit was 1.4 mg/kg. According to the authors this procedureis simple, accurate and very sensitive for honey, which allowsits use as a honey quality control method. However, in the paperneither precision nor recovery results were given. The authorsconcluded that natural formic acid content in honey is very vari-able and is influenced by honey botanical origin.

Ferreres et al. (11) isolated, identified, and quantified cis,trans-abscisic acid and trans, trans-abscisic acid in Portuguesebell heather (Erica sp.) honey by using reverse phase HPLCwith diode array detection. To confirm the structures, EIMS, 1HNMR and 13C NMR techniques were used. These authors didnot find these two acids in any other type of unifloral honeyanalysed, so they could be useful markers of Erica sp. honeys,after the analysis of a higher number of samples from differ-ent geographical origins. In a recent study, Gheldof et al. (5)found cis, trans-abscisic acid in honeys from different botan-ical origin (buckwheat, soy, tupelo, clover, fireweed, acacia),in quantities approximately 10 times less than the quantitiesfound by Ferreres et al. (11). Later, abscisic acids (cis, trans-abscisic acid and trans, trans-abscisic acid) were found in largeamounts in both Australian and New Zealand Leptospermunhoneys (46). In this work, the level of trans-trans abscisic acidin New Zealand manuka honey were much higher than that inAustralian jelly bush honey. Yao et al. (47) also investigated thepresence of these acids in Australian eucalyptus honeys. In stud-ies, cis, trans-abscisic acid and trans, trans-abscisic acid couldbe used individually and/or jointly for the authentication of thebotanical origins of these kind of honeys.

Del Nozal et al. (12) quantified citric, pyruvic, galacturonic,gluconic, malic, citramalic, quinic, succinic, fumaric and formicacids in 57 honeys from different botanical sources. Sampleswere previously purified by solid phase extraction (SPE) with

anionic exchange cartridges. These authors employed two sys-tems of HPLC with UV detection. One of the procedures wasbased on the use of an anionic exchange column (specific fororganic acids) with sulphuric acid 5 mM as mobile phase, at aflow rate of 0.5 mL/min. The other procedure used two reversedphase C18 columns connected in series, by employing ammo-nium dihydrogen phosphate pH = 2.2 as mobile phase, at a flowrate of 0.7 mL/min. With both methods the approximate analysistime was 30 minutes. In this paper, no result of precision andrecovery in honey samples is shown. Data of both precision andrecovery were given on the basis of organic acids standards orsyrups that tried to imitate honey composition.

Alamanni et al. (27) analyzed oxalic, lactic and formic acidscontents in 49 unifloral Sardinian (Italy) samples. Samples werepreviously purified by solid phase extraction (SPE) with an-ionic exchange cartridges and subsequently analyzed by HPLCwith UV detection. Chromatographic separation took place in20 minutes. These authors used a specific column for organicacids; mobile phase was sulphuric acid 0.025 N, at a flow rate of0.6 mL/min. Repeatability coefficients of variation were 2.8%for oxalic acid, 1.3% for lactic acid and 4.3% for formic acid.Average recovery percentages were 91.3% (oxalic acid), 94.5%(lactic acid) and 93.7% (formic acid). Detection limits rangedfrom 0.96 mg/L (formic acid) and 6.37 mg/L (lactic acid) andquantification limits ranged from 3.19 mg/L (formic acid) and21.22 mg/L (lactic acid). As it has been already commented,oxalic acid results were compared with those obtained by theenzymatic analysis, showing a good correlation between bothprocedures. In this work, the average natural oxalic acid con-tent was evaluated in honeys from different botanical sources toobserve if eventual treatment in beehives with oxalic acid deter-mined substancial or abnormal increases of the acid in honey.No increases were found when oxalic acid was used for thetreatment of Varroa infestation.

Del Nozal et al. (48) determined oxalic acid (as oxalate), to-gether with other inorganic anions such as sulphate and nitrate,in 99 floral and honeydew honeys by ionic chromatography (IC)with conductivity detection. To avoid matrix interferences, sam-ples were purified by solid phase extraction (SPE) with ionicexchange cartridges, procedure already set up by these authors.Chromatographic separation was carried out, in about 50 min-utes with an anionic exchange column and a mixture of lithiumborate-gluconate (1.4%), n-butanol (2.0%), acetonitrile (12.0%)and water (84.6%) as mobile phase at a flow-rate of 2 mL/min.Repeatability for oxalic acid was 4.05%, whereas average re-covery percentage was 92.72 ± 3.34%. The content of oxalicacid was evaluated in honey samples before and after a treatmentwith oxalic acid for varroasis control. The conclusion was thatthis therapeutical treatment of the beehives did not influence theinitial value of oxalic acid in honey.

Casella and Gatta (49) determined organic acids such as gal-lic, ascorbic, gluconic, lactobionic, galacturonic, and glucuronicacids by an anionic exchange chromatographic. The detec-tion was amperometric with constant voltage. Chromatographic

8 I. MATO ET AL.

separation took place in less than 20 minutes with an anionicexchange column, using an alkaline mobile phase composedby NaOH 0.1 M and sodium acetate 0.08 M. This methodwas applied to different samples such as tea, vinegar or honey.For honey, only gluconic acid was quantified, with a coeffi-cient of variation between 6% and 9% for the precision and98% for the recovery. According to these authors, their pro-cedure allowed determining various organic acids in real ma-trices avoiding complicate extraction and derivatization proce-dures and providing with good sensitivity, reproducibility andrecovery.

Suarez-Luque et al. (50, 51) set up a HPLC method with UVdetection that allowed determining malic, maleic, citric, succinicand fumaric acids in 15 minutes. Chromatographic separationwas carried out on a single C18 reversed phase column, usingas mobile phase Milli-Q water acidified with m-phosphoric acid4.5% (p/v) at pH 2.20 and using a flow rate of 0.7 mL/min. In or-der to avoid matrix interferences, sample was previously purifiedby solid phase extraction (SPE) with ionic exchange cartridges.These authors studied repeatabilities (%RSD ≤ 3.20% for thedifferent acids), reproducibilities (%RSD ≤ 4.86% for the dif-ferent acids) and recoveries (percentages that ranged between62.9% for malic acid and 99.4% for citric acid).

Del Nozal et al. (52, 53) separated oxalic, D-glucuronic, cit-ric, galacturonic, propionic, pyruvic, malic, citramalic, quinic,D-gluconic, lactic, formic, glutaric, fumaric, succinic and butyricacids in 39 honeys (52) and 58 honeys (53), all of them from thesame geographical origin and different botanical sources. Theseresearchers set up a HPLC, with UV (210 nm) detection proce-dure, by using four ion-exclusion columns connected in series,employing water with an 0.1% (v/v) of o-phosphoric acid asa mobile phase. The separation time was almost 60 minutes.To determine organic acids previous purification of the honeysamples was not necessary, because only dilution and later fil-tration were needed. With these conditions oxalic, D-glucuronic,citric, galacturonic, propionic, quinic, D-gluconic, formic, andglutaric acids were quantified. Injection repeatability studies aswell as complete repeatability studies of the method with bothstandards and honey were carried out. As regards honey, au-thors only gave results of oxalic, D-glucuronic, citric, galac-turonic, pyruvic, D-gluconic, formic and glutaric acids. RSD%was ≤5.5% for the repeatability of the injection and ≤7.2% forthe repeatability of all the procedure. In the paper, no resultof accuracy was given. To evaluate only the oxalic acid, twocolumns coupled in series were enough. In this case, analysistime was reduced to 10–15 minutes. Concentrations of the or-ganic acids studied were very variable, and even although theycould be influenced by the botanical origin of the samples, withineach group, an important variability of concentrations wasobserved.

Liquid chromatography is a technique widely used by lots ofresearchers to determine organic acids in many foodstuffs dueto its sensitivity, versatility and reproducibility. Nevertheless thequantities of honey’s organic acids are very low and thereby,

there are many compounds that interfere in the quantificationof organic acids, as for example, sugars. Interferences must beremoved by means of a pretreatment of the sample or by em-ploying columns in series, thus making these methods tediousand time consuming.

ELECTROPHORETIC METHODSOrganic acids separation and quantification is a very im-

portant application of capillary electrophoresis (54). Thereare lots of papers in which organic acids of several food-stuffs have been quantified by this method, many of thesepapers being review articles (55–59). Surprisingly, there arefew papers of honey’s organic acids determination by capillaryelectrophoresis.

Boden et al. (60) only identified honey’s citric acid andthe inorganic ions chloride and phosphate. Separation wascarried out in 8 minutes at 22◦C, with indirect-UV detec-tion at 232 nm, using as electrolyte salicylic acid 7.5 mM,TRIS (tris[hydroximethyl]aminomethane) 15 mM, Ca(OH)2

1 mM and DOTAOH (dodecyltrimethylammonium hydroxide)0.35 mM.

Navarrete et al. (61) applied a CZE method for the determi-nation of oxalic, fumaric, maleic and malic acids in honey. Theseparation has been carried out in 14 minutes. The composi-tion of running buffer was 50 mM sodium phosphate, 25% 2-propanol and 0.001% HDB (hexadimethrine bromide) adjustedto pH 8 with 1.0 NaOH. The instrumental parameters have beenalso optimized (voltage of −20 kV, hydrodynamic injection for15 seconds and UV detection at 210 nm). Repeatability andreproducibility studies were not carried out and the recoverypercentages ranged between 97.62% and 101.81%. Therefore,several important organic acids, as for example gluconic acidwhich is the predominant acid in honey or others, as citric, acetic,formic or lactic acids were not determined.

Mato et al. (62) developed a simple capillary electrophoresismethod with direct-UV detection for the determination of oxalic,formic, malic, succinic, pyruvic, acetic, lactic, citric and glu-conic acids in less than 4 minutes. The electrolyte compositionwas phosphate (7.5 mM NaH2PO4 and 2.5 mM Na2HPO4) as thecarrier buffer, 2.5 mM TTAOH (tetradecyltrimethylammoniumhydroxide, commercial name OFM-OH) as electroosmotic flowmodifier and 0.24 mM CaCl2 as selectivity modifier, adjustingthe pH at 6.40 constant value. The running voltage was −25 kVat controlled temperature of 25◦C. The injection was carried outin hydrodynamic mode and the detection mode was UV directat 185 nm. Only a simple treatment of dilution and filtrationwas necessary for the determination of organic acids in honeyby this electrophoretic method. Repeatability and reproducibil-ity studies were carried out and showed RSD (%) ≤4.6% and≤10.0 %, respectively. Recovery percentages ranged between95% and 103%.

Despite capillary electrophoresis has been hardly appliedto honey’s organic acids determination, seems to show lots of

METHODS OF ORGANIC ACIDS IN HONEY 9

advantages if compared with other procedures. In general, thistechnique is simple, rapid and low cost, because it needs neitherlaborious treatment of the samples nor long times of analysis.Furthermore, capillary electrophoresis avoids the most prob-lematic interferences found by applying HPLC, such as sugars,which are the main honey constituents (which represent 99%total solids). Capillary electrophoresis main disadvantage is itslower reproducibility, if compared with other methods, and itslittle lower sensitivity as well, because limits of detection andquantification are in general, higher than those obtained by enzy-matic and chromatographic procedures. Nevertheless, capillaryelectrophoresis is one of the techniques with better prospects forhoney’s organic acids determination, because it has been hardlydeveloped for this purpose, unlike its development regardingother foods.

CONCLUSIONFirstly, it is necessary to highlight that there are few litera-

ture references as regards honey’s organic acids determination.Nevertheless, due to the importance of these compounds, bothtraditional and new analytical procedures are very likely to bewidely studied and developed within future years.

Enzymatic analysis should be used when only one organicacid must be determined, because this method is very specificand provides excellent precision and accuracy.

Gas chromatography should be the chosen technique forvolatile or semi-volatile organic acids. This technique, to-gether with liquid chromatography and capillary electrophore-sis, provides the advantage of determining numerous organicacids. Despite being a versatile, sensitive and reproducible tech-nique, HPLC is tedious and time consuming because it re-quires previous purification of the samples and/or the use ofseveral columns in series to improve the separation of organicacids.

Capillary electrophoresis is a technique that takes very shorttime. In addition, sample preparation is simple. On the con-trary, capillary electrophoresis is less sensitive and more diffi-cult to reproduce than other techniques. Nevertheless, the de-velopment of capillary electrophoresis procedures to determineorganic acids in honey is extremely interesting, in particular dueto the fact that presently there are few papers published on thistopic.

REFERENCES1. J. W. Jr. White, Composicion y propiedades de la miel, in La Api-

cultura en los Estados Unidos, ed. S. E. McGregor (Mexico D. F.:Limusa, 1979), 57–66.

2. E. Crane, The traditional hive products: honey and beeswax, inBees and Beekeeping. Science, Practice and World Resources, ed.(Oxford: Heinemann Newnes, 1990), 388–451.

3. S. Bogdanov, Nature and origin of the antibacterial substances inhoney. Lebensm.-Wis. u.-Technol. 30 (1997):748–753.

4. R. J. Weston, K. R. Mitchell, and K. L. Allen, Antibacterial phe-nolic components of New Zealand manuka honey. Food Chemistry64 (1998):295–301.

5. N. Gheldof, X. -H. Wang, and N. J. Engeseth, Identification andquantification of antioxidant components of honeys from variousfloral sources. Journal of Agricultural and Food Chemistry 50(2002):5870–5877.

6. H. M. Eshbah and F. F. Ahmed, Screening certain natural productsand organic acids for controlling Varroa mite (Varroa jacobsoniOudemans). Shashpa 1 (1994):53–62.

7. F. Mutinelli, A. Baggio, F. Capolongo, R. Piro, L. Prandin, and L.Biasion, A scientific note on oxalic acid by topical application forthe control of varroosis. Apidologie 28 (1997):461–462.

8. N. W. Calderone, Effective fall treatment of Varroa jacobsoni(Acari: Varroidae) with a new formulation of formic acid incolonies of Apis mellifera (Hymenoptera: Apidae) in the North-eastern United States. Journal of Economic Entomology 93(2000):1065–1075.

9. A. Cherchi, L. Spanedda, C. Tuberoso, and P. Cabras, Solid-phaseextraction and high-performance liquid chromatographic determi-nation of organic acids in honey. Journal of Chromatography A669 (1994):59–64.

10. A. L. Wilkins, Y. Lu, and S.-T. Tan, Extractives from New Zealandhoneys. 5. Aliphatic dicarboxylic acids in New Zealand Rewerewa(Knightea excelsa) honey. Journal of Agricultural and Food Chem-istry 43 (1995):3021–3025.

11. F. Ferreres, P. Andrade, and F. A. Tomas-Barberan, Natural oc-currence of abscisic acid in heather honey and floral nectar.Journal of Agricultural and Food Chemistry 44 (1996):2053–2056.

12. M. J. Del Nozal, J. L. Bernal, P. Marinero, J. C. Diego, J. I. Frechilla,M. Higes, and J. Llorente, High performance liquid chromato-graphic determination of organic acids in honey from differentbotanical origin. Journal of Liquid Chromatography and RelatedTechnologies 21 (20) (1998):3197–3214.

13. I. Mato, J. F. Huidobro, J. Simal-Lozano, and M. T. Sancho, Sig-nificance of nonaromatic organic acids in honey. Journal of FoodProtection 66 (2003):2371–2376.

14. Methods of Enzymatic BioAnalysis and Food Analysis UsingTest-Combinations (Mannheim, Germany: Boehringer MannheimGmbH, 1995).

15. M. L. Tourn, A. Lombard, F. Belliardo, and M. Buffa, Quantitativeanalysis of carbohydrates and organic acids in honeydew, honeyand royal jelly by enzymic methods. Journal of Apicultural Re-search 19 (1980):144–146.

16. W. Stoya, G. Wachendoerfer, I. Kary, P. Siebentritt, and E.Kaiser, Formic acid as a therapeutic against varroatose and its ef-fect on honey. Deutsche Lebensmittel-Rundschau 82 (1986):217–221.

17. W. Stoya, G. Wachendoerfer, I. Kary, P. Siebentritt, and E.Kaiser, Milchsaure als Therapeutikum gegen Varroatose und ihreAuswirkung auf den Honig. Deutsche Lebensmittel-Rundschau 83(1987):283–286.

18. H. Hansen and M. Guldborg, Residues in honey and wax aftertreatment of bee colonies with formic acid. Tidsskr. Planteavl. 92(1988):7–10.

19. B. Talpay, Inhaltsstoffe des Honigs-Citronensaure (Citrat).Deutsche Lebensmittel-Rundschau 84 (1988):41–44.

20. B. Talpay, Inhaltsstoffe des Honigs-Ameisensaure (Formiat).Deutsche Lebensmittel-Rundschau 85 (1989):143–147.

21. A. G. Sabatini, G. L. Marcazzan, R. Colombo, and M. Garagnani,Applicazione di un metodo enzimatico per la determinazione

10 I. MATO ET AL.

dell’acido formico e dell’acido lattico presenti nel miele. Api-coltura 9 (1994):135–145.

22. I. Mato, J. F. Huidobro, M. P. Sanchez, S. Muniategui, M. A.Fernandez-Muino, and M. T. Sancho, Enzymatic determination oftotal D-gluconic acid in honey. Journal of Agricultural and FoodChemistry 45 (1997):3550–3553.

23. P. Pulcini, M. G. Piazza, and F. Allegrini, Total gluconic acid(AGT) content in Italian unifloral honeys. Industrie AlimentariXLIII (2004):263–268.

24. I. Mato, J. F. Huidobro, M. P. Sanchez, S. Muniategui, M.A. Fernandez-Muino, and M. T. Sancho, Enzymatic determina-tion of L-malic acid in honey. Food Chemistry 62 (1998):503–508.

25. I. Mato, J. F. Huidobro, V. Cendon, S. Muniategui, M. A.Fernandez-Muino, and M. T. Sancho, Enzymatic determina-tion of citric acid in honey by using polyvinylpolypyrrolidone-clarification. Journal of Agricultural and Food Chemistry 46(1998):141–144.

26. M. Cossu and Alamanni, Possibilita di impiego di una metodicaenzimatica per la determinazione dell’acido ossalico in campionidi miele della Sardegna. La Rivista di Scienza dell’Alimentazione28 (3) (1999):315–319.

27. M. C. Alamanni, M. Cossu, and F. Sanna, Un metodo HPLCper il dosaggio degli acidi ossalico, lattico e formico quali aca-ricidi e componenti naturali nel miele. La Rivista di Scienzadell’Alimentazione 29 (2000):139–147.

28. S. Bogdanov, J.-D. Charriere, A. Imdorf, V. Kilchenmann, and P.Fluri, Determination of residues in honey after treatments withformic and oxalic acid under field conditions. Apidologie 33(2002):399–409.

29. E. E. Stinson, M. H. Subers, J. Petty, and J. W. J. White, Thecomposition of honey. V. Separation and identification of the or-ganic acids. Archives of Biochemistry and Biophysics 89 (1960): 6–12.

30. V. K. Speer and A. Montag, Beitrag zum Vorkommen von Ben-zoesaure und Phenylessigsaure in Honig. Deutsche Lebensmittel-Rundschau 80 (1984):103–105.

31. V. K. Speer and A. Montag, Phenylacetic acid and benzoic acid con-tents in honey. Lebensmittelchem. Gerichtl. Chem. 39 (1985):11–12.

32. E. Steeg and A. Montag, Nachweis aromatischer carbonsaurenin honig. Aromatic carbonic acids of honey. Z. Lebensm. Unters.Forsch. 184 (1987):17–19.

33. E. Steeg and A. Montag, Quantitative Bestimmung aromatis-cher Carbonsauren in Honig (Quantitative determination of aro-matic carbonic acids in honey). Z. Lebensm. Unters. Forsch. 187(1988):115–120.

34. E. Steeg and A. Montag, Minorbestandteile des Honigs mit Aroma-Relevanz. I. Aromatische Carbonsauren und Ester aromatischerCarbonsauren. Deutsche Lebensmittel-Rundschau 84 (1988):103–108.

35. T. Echigo and T. Takenaka, Production of organic acids in honeyby honeybees. Nippon Nogei Kagaku Kaishi 48 (1974):225–230.

36. S.-T. Tan, P. T. Holland, A. L. Wilkins, and P. C. Molan, Extractivesfrom New Zealand honeys. 1. White Clover, Manuka, and KanukaUnifloral honeys. Journal of Agricultural and Food Chemistry 36(1988):453–460.

37. K. Horvath and I. Molnar-Perl, Simultaneous GC-MS quantitationof o-phosphoric, aliphatic and aromatic carboxylic acids, proline,hydroxymethylfurfurol and sugars as their TMS derivatives: in hon-eys. Chromatographia 48 (1998):120–126.

38. M. L. Sanz, M. Gonzalez, C. de Lorenzo, J. Sanz, and I.Martınez-Castro, A contribution to the differentiation between nec-tar honey and honeydew honey. Food Chemistry 91 (2005):313–317.

39. A. Verzera, S. Campisi, M. Zappala and I. Bonaccorsi, SPME-GC-MS analysis of honey volatile components for the characterizationof different floral origin. American Laboratory 33 (2001):18–21.

40. D. Pilz-Guther and K. Speer, Development of a GC method for thesimultaneous determination of organic acids in honey. DeutscheLebensmittel-Rundschau 100 (2004):84–87.

41. M. Perez-Cerrada, M. A. Herrero-Villen, and A. Maquieira, Sugar-rich food: Determination of inorganic anions by ionic chromatog-raphy. Food Chemistry 34 (1989):285–294.

42. E. Jorg and G. Sontag, Determination of phenolic acids in honeyby HPLC using coulometric dual electrode detection. DeutscheLebensmittel-Rundschau 88 (1992):179–183.

43. A. Cherchi, M. Porcu, L. Spanedda, C. I. G. Tuberoso, S. Cosentino,and F. Palmas, Individuazione di parametri utili per la caratteriz-zazione e la valorizzazione di mieli tipici della Sardegna: asfodelo,cardo e corbezzolo. La Rivista di Scienza dell’Alimentazione 24(1995):523–534.

44. J. Serra-Bonvehı, A. Bentabol-Manzanares, and J. M. Santos-Vilar,Quality evaluation of broom honey (Spartocytisus supranubius L)produced in Tenerife (The Canary Islands). Journal of the Scienceof Food and Agriculture 84 (2004):1097–1104.

45. A. Defilippi, G. Piancone, A. Prandtatter, and G. P. Tibaldi, Honeyquality: Ion chromatographyc determination of formic acid. Indus-trie Alimentari 34 (33) (1995):495–497, 503.

46. L. Yao, N. Datta, F. A. Tomas-Barberan, F. Ferreres, I. Martos, andR. Singanusong, Flavonoids, phenolic acids and abscisic acid inAustralian and New Zealand Leptospermum honeys. Food Chem-istry 81 (2003):159–168.

47. L. Yao, Y. Jiang, R. Singanusong, N. Datta, and K. Raymont,Phenolic acids and abscisic acid in Australian Eucalyptus hon-eys and their potential for floral authentication. Food Chemistry 86(2004):169–177.

48. M. J. Del Nozal, J. L. Bernal, J. C. Diego, L. A. Gomez, J. M.Ruiz, and M. Higes, Determination of oxalate, sulfate and nitratein honey and honeydew by ion-chromatography. Journal of Chro-matography A 881 (2000):629–638.

49. I. G. Casella and M. Gattta, Determination of electroactive organicacids by anion-exchange chromatography using a copper mod-ified electrode. Journal of Chromatography A 912 (2001):223–233.

50. S. Suarez-Luque, I. Mato, J. F. Huidobro, J. Simal-Lozano, andM. T. Sancho, Rapid determination of minoritary organic acidsin honey by high performance liquid chromatography. Journal ofChromatography A 955 (2002):207–214.

51. S. Suarez-Luque, I. Mato, J. F. Huidobro, and J. Simal-Lozano,Solid-phase extraction procedure to remove organic acids fromhoney. Journal of Chromatography B 770 (2002):77–82.

52. M. J. Del Nozal, J. L. Bernal, J. C. Diego, L. A. Gomez, andM. Higes, HPLC determination of low molecular weight or-ganic acids in honey with series-coupled ion-exclusion columns.

METHODS OF ORGANIC ACIDS IN HONEY 11

Journal of Liquid Chromatography and Related Technologies 26(2003):1231–1253.

53. M. J. Del Nozal, J. L. Bernal, L. A. Gomez, M. Higes, and A.Meana, Determination of oxalic acid and other organic acids inhoney and in some anatomic structures of bees. Apidologie 34(2003):181–188.

54. F. S. Stover, Organic acids and organic ions, in Handbook of Capil-lary Electrophoresis Applications ed. H. Shintani and J. Polonsky(London: Chapman and Hall, 1997), 550–579.

55. J. Lindeberg, Capillary electrophoresis in food analysis. FoodChemistry 55 (1996):73–94.

56. J. Lindeberg, Addendum to “Capillary electrphoresis in food anal-ysis.” Food Chemistry 55 (1996):95–101.

57. C. W. Klampfl, W. Buchberger, and P. R. Haddad, Determinationof organic acids in food samples by capillary zone electrophoresis.Journal of Chromatography A 881 (2000):357–364.

58. R. A. Frazier, Recent advances in capillary electrophoresis methodsfor food analysis. Electrophoresis 22 (2001):4197–4206.

59. V. Galli, A. Garcıa, L. Saavedra, and C. Barbas, Capillary elec-trophoresis for short-chain organic acids and inorganic anions indifferent samples. Electrophoresis 24 (2003):1951–1981.

60. J. Boden, I. Haumann, and A. Mainka, Anwendung der Kap-illarelektrophorese in der Lebensmittelanalytik. GIT Labor-Fachzeitschrift. 44 (2000):924–927.

61. M. Navarrete, S. Casado, M. Minelli, A. Segura, A. Bonetti, G.Dinelli, and A. Fernandez, Direct determination of aliphatic acidsin honey by coelectroosmotic capillary zone electrophoresis. Jour-nal of Apicultural Research 44 (2005):65–70.

62. I. Mato, J. F. Huidobro, J. Simal-Lozano, and M. T. Sancho, Simul-taneous determination of nonaromatic organic acids in honey bycapillary zone electrophoresis with direct UV detection. Journalof Agricultural and Food Chemistry (Submitted for publication).